© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 66202
    Prev
    Next
    INDEX
    Explanations

    released topics

    np_acts-logits-general · gemini-2.5-flash-lite

    This neuron fires on CJK tokens containing the character “出” (as in Japanese “出力” or Chinese “出了”), i.e. verbs or noun-verbs built around “出.”

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    "
    -3.52
    }$
    -3.47
    ;
    -3.11
    u
    -2.80
    [
    -2.66
    >
    -2.61
    man
    -2.56
    +
    -2.50
    :
    -2.50
    </td>
    -2.50
    POSITIVE LOGITS
    Ꮞ
    3.27
    焿
    3.19
    掼
    3.06
    乆
    3.06
    鑪
    2.97
    
    2.95
    NOSIS
    2.91
    bluza
    2.84
    メリカ
    2.83
    erintah
    2.81
    Activations Density 0.001%

    No Known Activations